SOFTWARE PATENTABILITY UPDATE

By Curtis L. Harrington

Editor's Note: Curt Harrington contributes some recent information on patents in his latest article for SSS. The world of intellectual property rights is changing very quickly. In the US alone, tremendous turmoil is underway in almost all technical field of endeavor involving the implications of patentability. Perhaps none is more controversial than the field of software.

The court keep pulling for change in the direction of patentability of software as the U.S. Patent and Trademark Office continues to resist such change. The resistance has extended to the shenanigan level as is exemplified in the recent embattled case; In re Alappat 31 USPQ2d 1545 (CAFC 1994).

This case was earlier reported as Ex parte Alappat 23 USPQ2d 1340 (Board of Patent Appeals 1992). The claims in the subject patent application were directed to a "device" for converting vectors in a data list. This invention was in essence a method for treating computer screen data to smooth out the displayed line trace. In the more recently issued case, it was revealed that the invention was primarily intended for use with an oscilloscope output shown on a cathode ray tube.

The claims, and particularly the broadest claim, was direct to a "rasterizer" and included structure such as "means for determining a vertical distance of vectors," "means for determining an elevation of a row of pixels spanned by the vector," "means for normalizing the vertical distance and elevation, "and, "means for outputting illumination intensity data as a function of the normalized vertical distance and elevation."

The physical devices used to perform these tasks included several arithmetic logic circuits, barrel shifters and a read-only memory, all digital computational devices.

The history of the case, however, provides the most interest and controversy. The above claim was first disallowed by the Examiner. On appeal to the Board of Interferences and Appeals, a three member panel reversed the Examiner and indicated the claims to be allowable. The Examiner, and probably at the direction of others higher up in the Patent Office, requested reconsideration. In response, an eight-member panel of the Board of Appeals was constituted and including the Patent and Trademark Office Commissioner, Deputy Commissioner, Assistant Commissioner, among others, mentioned to emphasize the top-heavy, policy-making composition of the board. The eight-member panel then promptly held the claim to be unpatentable, a move which has been popularly referred to as a reversal more bases upon the preferences of the Patent and Trademark Office management and policy rather than an objective examination of the merits of a particular case.

The CAFC, with due discussion and agreement of the legality of having the case pulled back into the eight-member board with the Commissioner sitting in judgment, promptly reversed the holding of the eight-member board and held the claim to be statutorily allowable subject matter.

Technically, arguments against patentability were that each "means for" portion of the claims was in fact a process step being performed by a conventional computer. The CAFC pointed out that not only did the opening paragraph claim a machine (rasterizer) but that since "means for claims are to be interpreted narrowly, as including the devices disclosed in the patent and their equivalents, that the claims are interpreted as having the devices in each portion of the claim.

But how does this result guide bussiness people to know what is now patentable and what is not patentable, especially with an ever expanding explosion in software and multi-media? Reading thus far it would appear that if the circuitry which was responsible for each step was identifiable or at least identifiably distinct from other portions of the computational circuitry, and was identified and claimed that way in a patent application, that this would be a step in the right direction toward getting the claim allowed. This means that one who seeks rights protection should know a great deal more detail about the device in question rather than its status as a "computer," if there is to be a realistic expectation of allowability.

Other comments in In re Alappat which give further guidance indicate that the invention should be "not a disembodied mathematical concept which may be characterized as an 'abstract idea,' but rather a specific machine to produce a useful, concrete, and tangible result." This seems to indicate that the more specialized the machine or computer is shown to be in the patent application, the greater the probability of it being fond to be patentable subjectmatter.

Another guide is directed to the practitioner and how he drafts his claims, with the further comment "Indeed, claim 15 as written is not so abstract that it would 'wholly pre-empt' the use of any apparatus employing the combination of mathematical calculations recited therein." Phrased another way, this is a proscription on the zeal in claiming broadly, or at least broadly enough to begin to take possession of the mathematical function for machines generally. Pushing the claim breadth to a limit which exceeds the particular application shown in the patent, but which falls arguably short of pre-empting use in all other devices will require delicate crafting of the claims and solid arguments in their defense.

Lastly the possibility of rejection base upon an argument that the claimed invention "reads on ( is capable of being performed by) a standard computer" was eliminated with the statement of a simple rule: that the claim "would read on a general purpose computer programmed to carry out the claimed invention ... alone does not justify holding [the claim] unpatentable as directed to nonstatutory subject matter." This comment tends to leave open the upper limit of the "wholly preempt" language of the previous paragraph.

The holding in In re Alappar fails to cut an exact dimension into what is patentable and what is not. Principles often practiced in writing patents for computer inventions have been bolstered by the court's holding. These principles have dictated (1) the inclusion of an analog version of the invention which parallels the digital version if it is available; (2) to the extent possible attempt to identify individual components according to their function, either by block diagram. or subdivision of function or location; and (3) attempt to incorporate as much structure into the claim as is possible without unduly limiting it.

Specifically avoiding any references to restrictions on the type of mathematical functions involved in a patent, the court is seeming to open the door to all varieties of computer inventions, provided that the patent practitioner sets them forth in a concrete manner, with a reasonable degree of differentiable identification of the elements which perform the different functions in practicing the invention, and where the claimant is Willing to limit the claims to a device or process which can be defined in a way which leaves the mathematics and at least some machine `uses of those mathematics to the public domain.

Since the holding in In re Alappat, the court has had an opportunity to again consider the statutory status of a computer process in In re Warmerdam (CAFC 1994) 31 USPQ2d 1754. Here the invention was a method claim for generation a data structure representing the shape of a physical object, an application useful in robotics. The claimed process performed with the steps of (1) locating the medial axis of the object, and (2) creating a hierarchy of [spatially dimensioned] bubbles on the medial axis, This process was for the purpose of creating a zone within which a robotic arm would be prohibited from travelling during its motion. Based on the earlier holding of In re Alappat, the Board of Patent Appeals rejection of this simple two step claim was upheld by the court. It seems to have violated the interpretive criteria set forth above in that there was very little differentiation of function, no significant infusion of physical stucture or steps, and only a modicum of limitation in requiring that the method take place in a position and/or motion control machine. The patent application claims also fell short of the critera that the claim relate to a specific machine to produce a specific result (no machine was recited in the actual performace of the claimed stepd).

Claim 5, however, which was drawn to "a machine having a memory which contains data representing a bubble hierachy generated by the method of any of claims 1-4)" and which was held rejected by the Patent Office Apeal Board for indefiniteness, was reversed and held allowable by the court. The rejection was based upon a lack of showing as to how a memory is produced by the steps of the process claims. The court held that since no one asserted that it would be difficult to produce the memory from the claims, that it is therefore assumed possible: and that the claim covered all of the ways of doing it.

It thus appears as if the much wider invitations to patent of In re Alappat are being exercised freely and supported by the subsequent holding in In re Warmerdam. These cases represent significant liberalizations. However, given the continued anti-computer patent stance of the U.S. Patent Office, as reflected in the manipulative re-hearing of the original Board of Appeals decision before the eight-member panel, the expanded protection to be afforded current software applicants will likely require aggressive prosecution and advocacy.

CURTIS L. HARRINGTON, is a partner with the Newport Beach law firm of Hawes & Fischer (714-759-6601) and practices in the areas of Patent, Trademark, Copyright and Trade Secret law. He is admitted to practice before the state bars of California, Texas, and Arizona, and is admitted to practice before the U. S. District Court, the U.S. Court of Appeals, Fifth and Ninth Circuits, the U.S. Supreme Court and registered to practice before U.S. Patent and Trademark Office. His education includes a B.S. Chemistry (Auburn University, 1974); M.S. Chemical Engineering (Georgia Institute of Technology, 1977); M.S. Electrical Engineering (California State University - Long Beach, 1990); a Master of Business Administration (University of Oklahoma, 1985); and Juris Doctorate (University of Houston, 1983). He is a member of the Long Beach Bar Association Board of Governors, the Orange County Bat Association and the Japan America Society of Southern California.

Return to Table of Contents for this issue

In October 2000, this website and the copyright to all editions of Spread Spectrum Scene Online was purchased by SSS Online, Inc., and is operated by Pegasus Technologies. For more of the best information on RF, Spread Spectrum and wireless, press one of the buttons below: